Validation of the Weather Research and Forecast Model to Infer Main Drivers of Sundowner Winds in Coastal Santa Barbara

Sundowner wind is a type of downslope windstorm that occur in the lee of the Santa Ynez Mountains in Southern California. This wind typically initiates early evening, and ceases mid-morning, and is characterized by rapid warming and a sudden decrease in relative humidity. Sundowner winds have played a major role in the development of all significant wildfires in the Santa Barbara County in Southern California. These winds occur year round and peak during spring. Despite the importance of these winds to the local and regional community, the main meso-to-microscale drivers of Sundowners remain unknown. The existing network of ground stations is insufficient to examine mechanisms associated with the mesoscale evolution of these winds. Therefore, to advance understanding and increase lead-time predictability of Sundowner winds it is imperative to assess the skill of mesoscale models in reproducing the main observed features, including onset and end of events, timing of maximum intensity and spatial variability. For future fire decision support a development of a simple forecasting tool is necessary based on modeling studies. The objective of this work is to evaluate the skill of the Weather Research & Forecasting model (WRF) at 1km resolution in simulating significant Sundowner events. These events are selected to represent distinct seasons and large-scale forcing. We also evaluate simulations representative of non-Sundowner conditions. Specifically, this study performs sensitivity tests of boundary layer schemes to identify the relative importance of the choice of parameterizations to represent the diurnal cycle of winds, temperatures and relative humidity during Sundowner and in quiescent (no Sundowner) periods. We validate the model by using a network of surface stations and a recently installed wind profiler (installed in 2016) to highlight the strengths and weaknesses of the model to capture the boundary layer structure and surface properties. Then we use the validated simulations to further investigate the main drivers of Sundowner events. Understanding the relevant mechanisms is of main importance to enhance the forecasting capability of Sundowner events which can aid in future wildfire prediction and prevention.